Abhijit Mishra's photo

Assistant Professor of Practice
School of Information
The University of Texas at Austin
1616 Guadalupe St, Suite #5.202
Austin, TX - 78701

LinkedIn | Google Scholar


About me | Research | Teaching | Publications | Talks | Patents | Services | Students I am an Assistant Professor of Practice at the University of Texas School of Information. I teach courses such as Applied Machine Learning, Natural Language Processing (NLP), Deep Learning, and Introduction to Human-Centered Data Science. Before serving as a faculty, I was a research scientist at Apple inc., Seattle and IBM India Research Lab working on language understanding and generation components of Siri and IBM Watson. I have obtained Ph.D. in Computer Science and Engineering from the Indian Institute of Technology Bombay. My primary area of interest is Machine Learning for Natural Langauge Processing and I am quite passionate about problems related to language generation, language models, and multimodal and cognition-inspired NLP. In my last role at Apple, I was responsible for developing multimodal and multilingual on-device models for understanding users' intent when they interact with Siri in a privacy-preserving manner. The work touches upon concepts such as multimodal language models, modality fusion, model distillation, and compression. Over the years, I have been quite fascinated by generative modeling for conversational AI and have spent a considerable amount of time working on problems related to Natural Language Generation (NLG), dealing with data-to-text and text-to-text generation paradigms. In data-to-text, my research aims at generating natural language descriptions from structured data such as knowledge graphs, tables, etc. In text-to-text, my focus has been on problems such as text simplification, text style transfer and controllable paraphrasing.
  1. Vinayak Sastri, Joydeep Mondal, Abhijit Mishra, Seema Nagar, Kuntal Dey. 2019. Dynamic Content Rating Assistant. 20210065043. US Patents and Trademarks Office (USPTO)
  2. Abhijit Mishra, Enara C Vijil, Seema Nagar, Kuntal Dey. 2018. Question Answering System influenced by User Behavior and Text Metadata Generation. 20200302316. US Patents and Trademarks Office (USPTO)
  3. Abhijit Mishra, Parag Jain, Anirban Laha, Karthik Sankaranarayanan. 2018. Generation of Variable Natural Language Descriptions From Structured Data. 20200073944. US Patents and Trademarks Office (USPTO)
  4. Parag Jain, Amar Azad, Abhit Mishra, Karthik Sankaranarayanan. 2018. Unsupervised Tunable Stylized Text Transformations. 20200034432. US Patents and Trademarks Office (USPTO)
  5. Abhijit Mishra, Anirban Laha, Parag Jain, Karthik Sankaranarayanan. 2018. Real time assessment of Text Consistency. 20200302011. US Patents and Trademarks Office (USPTO)
  6. Abhijit Mishra, Parag Jain, Amar Azad, Karthik Sankaranarayanan. 2018. Controllable Style-Based Text Transformation20200311195, US Patents and Trademarks Office (USPTO)
Senior Program Committee Member Program Committee Member / Reviewer Organizing Committee member UT School of Information Students
  1. Rosalyn Lu (Informatics Honors Thesis, 2024, Ongoing): Multilingual Multimodal VLMs
  2. Thang Troung (Informatics Honors Thesis, 2024, Ongoing): Multilingual Multimodal VLMs
  3. Walt Wu (Informatics Capstone, 2024, Completed): Synthetic Data Generation for Query Rewriting
  4. Sean Fu (Informatics Capstone, 2024, Completed): Multimodal SLMs for Query Rewriting
  5. Krishna Shri Somepalli (Master's Report, 2024, Graduated): Personal Identifiable Information Discovery using Language Models
  6. Shreya Shukla (Master's Report, 2024, Graduated): EEG to Text Generation
  7. Mingda Li (Informatics Capstone, 2023, Graduated): Zero Short Visual Knowledge Question Answering with LLMs
  8. Carla Gonzalez (Informatics Capstone, 2023, Graduated): Concept Guided LLMs for Zero Short Visual Knowledge Question Answering (2023 Informatics Capstone)
  9. Sonali Hornick (Informatics Capstone, 2023, Graduated): An Empirical Evaluation of LLMs for Zero-shot Visual Knowledge Question Answering
Co-supervising Outside of UT
  1. Kishan Maharaj (MS Research, IIT Bombay, Ongoing): Cognitively Inspired NLP for Hallucination Detection in Generative NLP Models
  2. Ashita Saxena (MS Research, IIT Bombay, Graduated): Hallucination Detection and Mitigation in Generative NLP Models
Internships Mentored
  1. Neha Hulkund (BS, MIT, 2021): Leveraging Scene Descriptions for Open Domain Visual Question Answering, at Apple inc., Summer 2021
  2. Ivy Zhang (Rotation Engineer, Apple inc., 2021): Adapting T5 Closed Book QA for Knowledge Oriented Visual Queries at Apple inc., Summer 2021
  3. Kevin Patel (Ph.D., IIT Bombay, 2019): Explaining Black-box NLP Models through Eye-tracking at IBM Research, Summer 2019
  4. Sai Surya (B.Tech, IIT Kharagpur, 2018): Unsupervised Neural Text Simplification at IBM Research, Summer 2018
  5. Krishna Guddipati (B.Tech, IIT Bombay, 2017): Scoring Grammaticality of NLG output at IBM Research, Summer 2017
Last updated on 19/07/2023 at 5:00 p.m. CST